1. Field of the Invention
This invention relates to storage subsystems, and more specifically to validation of archived data at storage subsystems.
2. Description of the Related Art
The storage of data has always been important in data systems and enterprises and businesses. During the normal course of business, many companies generate massive amounts of information and data, much of which must be stored. Many times, this data must not only be stored, but must be stored and maintained for extended periods of time. Therefore, data such as this is usually archived in storage for possible retrieval at a later date. Archived data is generally stored on a backup disk(s) or storage device(s), and not on what may be considered as a primary disk(s) or storage device(s).
Archiving large amounts of data requires large amounts of storage and, therefore, the storage requirements can be costly. However, disk bit cost for storage devices such as ATA disk have currently become similar to that of tape or optical medium bit cost. Thus, many companies and vendors are considering using current disk subsystems as archived storage subsystems. However, disk subsystems that have been used as a primary storage or disk, many times lacks the capabilities desired for an archive storage subsystem. Over time, it is possible that some corruption or errors may have occurred on some or all of the archived data. Therefore, since data may have been stored for an extended period of time, upon retrieval of this archived data, validation or verification as to the correctness of the data may be desired. This process is currently performed by servers, which are connected to the storage subsystem. In systems where data is constantly archived by a server onto a storage subsystem, as the data archived increases year by year, an archive server will have to spend a lot of time to verify or validate the correctness of retrieved data.
Currently, many systems use redundant arrays of inexpensive disks (RAID) storage subsystems for data validation. RAID systems may be able to detect the corruption of data, but currently RAID systems perform validation of data using a server connected to the RAID storage subsystem. Data validation occurs at the server after the data is read from the storage subsystem. However, data verification of data on a storage subsystem at a server is problematic in that this requires a heavy overhead of substantial I/O between the server and the storage subsystem to verify all the data. Moreover, the server must perform many calculations to ensure validation of the data. This can have a negative effect on other archived data retrieving tasks.